On cluster analysis of complex and heterogeneous data

3rd SMTDA Conference Proceedings, 11-14 June 2014, Lisbon Portugal.

Agrupamento de dados visual interactivo

Com a crescente geração, armazenamento e disseminação da informação nos últimos anos, o anterior problema de falta de informação transformou-se num problema de extracção do conhecimento útil a partir da informação disponível. As representações visuais da informação abstracta têm sido utilizadas para auxiliar a interpretação os dados e para revelar padrões de outra forma escondidos. A visualização de informação procura aumentar a cognição humana aproveitando as capacidades visuais humanas, de forma a tornar perceptível a informação abstracta, fornecendo os meios necessários para que um humano possa absorver quantidades crescentes de informação, com as suas capacidades de percepção. O objectivo das técnicas de agrupamento de dados consiste na divisão de um conjunto de dados em vários grupos, em que dados semelhantes são colocados no mesmo grupo e dados dissemelhantes em grupos diferentes. Mais especificamente, o agrupamento de dados com restrições tem o intuito de incorporar conhecimento a priori no processo de agrupamento de dados, com o objectivo de aumentar a qualidade do agrupamento de dados e, simultaneamente, encontrar soluções apropriadas a tarefas e interesses específicos. Nesta dissertação é estudado a abordagem de Agrupamento de Dados Visual Interactivo que permite ao utilizador, através da interacção com uma representação visual da informação, incorporar o seu conhecimento prévio acerca do domínio de dados, de forma a influenciar o agrupamento resultante para satisfazer os seus objectivos. Esta abordagem combina e estende técnicas de visualização interactiva de informação, desenho de grafos de forças direccionadas e agrupamento de dados com restrições. Com o propósito de avaliar o desempenho de diferentes estratégias de interacção com o utilizador, são efectuados estudos comparativos utilizando conjuntos de dados sintéticos e reais.

Shear-induced lamellar phase of an ionic liquid crystal at room temperature

The phase behaviour of a number of N-alkylimidazolium salts was studied using polarizing optical microscopy, differential scanning calorimetry and X-ray diffraction. Two of these compounds exhibit lamellar mesophases at temperatures above 50 degrees C. In these systems, the liquid crystalline behaviour may be induced at room temperature by shear. Sheared films of these materials, observed between crossed polarisers, have a morphology that is typical of (wet) liquid foams: they partition into dark domains separated by brighter (birefringent) walls, which are approximately arcs of circle and meet at "Plateau borders" with three or more sides. Where walls meet three at a time, they do so at approximately 120 degrees angles. These patterns coarsen with time and both T1 and T2 processes have been observed, as in foams. The time evolution of domains is also consistent with von Neumann's law. We conjecture that the bright walls are regions of high concentration of defects produced by shear, and that the system is dominated by the interfacial tension between these walls and the uniform domains. The control of self-organised monodomains, as observed in these systems, is expected to play an important role in potential applications.

How foam-like is the shear-induced lamellar phase of an ionic liquid crystal?

Philosophical Magazine Letters Volume 88, Issue 9-10, 2008 Special Issue: Solid and Liquid Foams. In commemoration of Manuel Amaral Fortes

Sorption of phenanthrene on agricultural soils

Polyaromatic hydrocarbon (PAH) sorption to soil is a key process deciding the transport and fate of PAH, and potential toxic impacts in the soil and groundwater ecosystems, for example in connection with atmospheric PAH deposition on soils. There are numerous studies on PAH sorption in relatively low organic porous media such as urban soils and groundwater sediments, but less attention has been given to cultivated soils. In this study, the phenanthrene partition coefficient, KD (liter per kilogram), was measured on 143 cultivated Danish soils (115 topsoils, 0–0.25-m soil depth and 28 subsoils, 0.25–1-m depth) by the single-point adsorption method. The organic carbon partition coefficient, KOC (liter per kilogram) for topsoils was found generally to fall between the KOC values estimated by the two most frequently used models for PAH partitioning, the Abdul et al. (Hazardous Waste & Hazardous Materials 4(3):211– 222, 1987) model and Karickhoff et al. (Water Research 13:241–248, 1979) model. A less-recognized model by Karickhoff (Chemosphere 10:833–846, 1981), yielding a KOC of 14,918 Lkg−1, closely corresponded to the average measured KOC value for the topsoils, and this model is therefore recommended for prediction of phenanthrene mobility in cultivated topsoils. For lower subsoils (0.25–1-m depth), the KOC values were closer to and mostly below the estimate by the Abdul et al. (Hazardous Waste & Hazardous Materials 4(3):211–222, 1987) model. This implies a different organic matter composition and higher PAH sorption strength in cultivated topsoils, likely due to management effects including more rapid carbon turnover. Finally, we applied the recent Dexter et al. (Geoderma 144:620–627, 2008) theorem, and calculated the complexed organic carbon and non-complexed organic carbon fractions (COC and NCOC, grams per gram). Multiple regression analyses showed that the NCOC-based phenanthrene partition coefficient (KNCOC) could be markedly higher than the COCbased partition coefficient (KCOC) for soils with a clay/OC ratio <10. This possibly higher PAH sorption affinity to the NCOC fraction needs further investigations to develop more realistic and accurate models for PAH mobility and effects in the environment, also with regard to colloid-facilitated PAH transport.

Exploring nature profits: development of novel and potent lipophilic antioxidants based on galloylecinnamic hybrids

Phenolic acids are ubiquitous antioxidants accounting for approximately one third of the phenolic compounds in our diet. Their importance was supported by epidemiological studies that suggest an inverse relationship between dietary intake of phenolic antioxidants and the occurrence of diseases, such as cancer and neurodegenerative disorders. However, until now, most of natural antioxidants have limited therapeutic success a fact that could be related with their limited distribution throughout the body and with the inherent difficulties to attain the target sites. The development of phenolic antioxidants based on a hybrid concept and structurally based on natural hydroxybenzoic (gallic acid) and hydroxycinnamic (caffeic acid) scaffolds seems to be a suitable solution to surpass the mentioned drawbacks. Galloylecinnamic hybrids were synthesized and their antioxidant activity as well as partition coefficients and redox potentials evaluated. The structureepropertyeactivity relationship (SPAR) study revealed the existence of a correlation between the redox potentials and antioxidant activity. The galloylecinnamic acid hybrid stands out as the best antioxidant supplementing the effect of a blend of gallic acid plus caffeic acid endorsing the hypothesis that the whole is greater than the sum of the parts. In addition, some hybrid compounds possess an appropriate lipophilicity allowing their application as chain-breaking antioxidant in biomembranes or other type of lipidic systems. Their predicted ADME properties are also in accordance with the general requirements for drug-like compounds. Accordingly, these phenolic hybrids can be seen as potential antioxidants for tackling the oxidative status linked to the neurodegenerative, inflammatory or cancer processes.

A compositional scheduling framework for digital avionics systems

ARINC specification 653-2 describes the interface between application software and underlying middleware in a distributed real-time avionics system. The real-time workload in this system comprises of partitions, where each partition consists of one or more processes. Processes incur blocking and preemption overheads and can communicate with other processes in the system. In this work we develop compositional techniques for automated scheduling of such partitions and processes. At present, system designers manually schedule partitions based on interactions they have with the partition vendors. This approach is not only time consuming, but can also result in under utilization of resources. In contrast, the technique proposed in this paper is a principled approach for scheduling ARINC-653 partitions and therefore should facilitate system integration.

Segmentation of striatal brain structures from high resolution pet images

Dissertation presented at the Faculty of Science and Technology of the New University of Lisbon in fulfillment of the requirements for the Masters degree in Electrical Engineering and Computers

Enhancing the selection of a model-based clustering with external categorical variables

In cluster analysis, it can be useful to interpret the partition built from the data in the light of external categorical variables which are not directly involved to cluster the data. An approach is proposed in the model-based clustering context to select a number of clusters which both fits the data well and takes advantage of the potential illustrative ability of the external variables. This approach makes use of the integrated joint likelihood of the data and the partitions at hand, namely the model-based partition and the partitions associated to the external variables. It is noteworthy that each mixture model is fitted by the maximum likelihood methodology to the data, excluding the external variables which are used to select a relevant mixture model only. Numerical experiments illustrate the promising behaviour of the derived criterion. © 2014 Springer-Verlag Berlin Heidelberg.

Optimização de uma máquina de pultrusão

Este trabalho teve como principal objectivo optimizar um equipamento de produção de perfis pultrudidos já existente na empresa ALTO – PERFIS PULTRUDIDOS, Lda. O trabalho surgiu na sequência de um Projecto financiado pelo Programa QREN – Quadro de Referência Estratégico Nacional, determinadas debilidades identificadas no processo de pultrusão, principalmente ao nível da eficiência térmica na fieira e de alguma falta de produtividade devida às diversas operações necessárias à mudança da fieira consoante o tipo de perfil a produzir. Após um levantamento prévio da situação e uma adequada segmentação da máquina nas diferentes partes que a constituem e que contribuem activamente para o processo de produção dos perfis, foi elaborada uma lista de prioridades e foram sendo procuradas as soluções mais adequadas para cada caso, sempre com a participação activa da empresa, com vista à sua implementação final. A metodologia adoptada passou sempre por uma reunião inicial com os representantes da empresa e com os orientadores, efectuando-se a tradicional “tempestade de ideias”. Depois da correspondente maturação, desenvolvimento e aprovação prévia, por parte da empresa, as ideias foram desenvolvidas e até implementadas na sua maioria. O saldo poderá considerar-se extremamente positivo, tanto para a empresa que, ao implementar as soluções, as validou e ganhou competitividade, como para as pessoas envolvidas neste projecto, através da enorme aprendizagem adquirida.

Probabilistic consensus clustering using evidence accumulation

Clustering ensemble methods produce a consensus partition of a set of data points by combining the results of a collection of base clustering algorithms. In the evidence accumulation clustering (EAC) paradigm, the clustering ensemble is transformed into a pairwise co-association matrix, thus avoiding the label correspondence problem, which is intrinsic to other clustering ensemble schemes. In this paper, we propose a consensus clustering approach based on the EAC paradigm, which is not limited to crisp partitions and fully exploits the nature of the co-association matrix. Our solution determines probabilistic assignments of data points to clusters by minimizing a Bregman divergence between the observed co-association frequencies and the corresponding co-occurrence probabilities expressed as functions of the unknown assignments. We additionally propose an optimization algorithm to find a solution under any double-convex Bregman divergence. Experiments on both synthetic and real benchmark data show the effectiveness of the proposed approach.

Petri nets based components within globally asynchronous locally synchronous systems

Dissertação apresentada na Faculdade de Ciências e Tecnologias da Universidade Nova de Lisboa para a obtenção do grau de Mestre em Engenharia Electrotécnica e Computadores

Enhancing the selection of a model-based clustering with external categorical variables

In cluster analysis, it can be useful to interpret the partition built from the data in the light of external categorical variables which are not directly involved to cluster the data. An approach is proposed in the model-based clustering context to select a number of clusters which both fits the data well and takes advantage of the potential illustrative ability of the external variables. This approach makes use of the integrated joint likelihood of the data and the partitions at hand, namely the model-based partition and the partitions associated to the external variables. It is noteworthy that each mixture model is fitted by the maximum likelihood methodology to the data, excluding the external variables which are used to select a relevant mixture model only. Numerical experiments illustrate the promising behaviour of the derived criterion.

Petri net model decomposition - a model based approach supporting distributed execution

Dissertação apresentada para obtenção do Grau de Doutor em Engenharia Electrotécnica, Especialidade de Sistemas Digitais, pela Universidade Nova de Lisboa, Faculdade de Ciências e Tecnologia

A MAP approach to evidence accumulation clustering

The Evidence Accumulation Clustering (EAC) paradigm is a clustering ensemble method which derives a consensus partition from a collection of base clusterings obtained using different algorithms. It collects from the partitions in the ensemble a set of pairwise observations about the co-occurrence of objects in a same cluster and it uses these co-occurrence statistics to derive a similarity matrix, referred to as co-association matrix. The Probabilistic Evidence Accumulation for Clustering Ensembles (PEACE) algorithm is a principled approach for the extraction of a consensus clustering from the observations encoded in the co-association matrix based on a probabilistic model for the co-association matrix parameterized by the unknown assignments of objects to clusters. In this paper we extend the PEACE algorithm by deriving a consensus solution according to a MAP approach with Dirichlet priors defined for the unknown probabilistic cluster assignments. In particular, we study the positive regularization effect of Dirichlet priors on the final consensus solution with both synthetic and real benchmark data.